Notes on the Panama Thrush-Warbler

'•Vol.1913 XXX' J I CLARK,The Panama Thrush-Warbler. 11

NOTES ON THE PANAMA THRUSH-WARBLER.

BY HUBERT LYMAN CLARK. 1

T•ANKS to the kindnessof Mr. Outram Bangs,I have recently had the opportunityof examiningsome fine alcoholicmaterial of the PanamaThrush-Warbler, Rhodinocichla rosea eximia Ridgway. Mr. Bangs called my attention to the fact that the systematic positionof this bird is still unsettledand he suggestedthat a study of someof the anatomicaldetails of its structuremight throw light on its relationships. For his kindnessin placing the material in my hands, without restrictions,and for the loan of skinsof several other generawhich I wished to examine, I desire to expresshere my hearty thanks to Mr. Bangs. I shouldbe rash indeedif I expectedto actually settle, by these notes, the systematic position of Rhodinocichlafor like many . anothergenus of Passerinebirds, this one approachesmore or less nearly severaldifferent families and with whichone it is mostclosely affiliated is largely a matter of opinion. All I hope to do is to point out some features of the anatomy not previously known, summarizethose which have beendescribed, and expressmy own opinionas to the relationshipwhich thesefacts seemto indicate. Bill. The bill is rather slender,about 18 min. long, 5 min. wide at baseand 8 mm. deepat the samepoint. The upper mandible is distinctly curved but the lower is remarkably straight. The tomia are entire with neither tooth nor notch, and the same is true of the edgesof the upper mandibleexcept near the tip, where there is a large, roundednotch. This notch is exactlylike that which is found in the sameposition in the bill of certain tanagers, Eucometis,Mitrospingus, etc. Indeed, the bill of Rhodinocichla is more like that of Mitrospingusthan like that of any other bird with which I have comparedit. Nostrils. The nostrils show no distinctive character. They are ellipsoidal,longer than high and quite bare; the skin back of

1 Read at the meeting of the American Ornithologists' Union, November 13, 1912. 12 CL•, ThePanama Thrush-Warbler. LJan.['Auk them is free from feathers for a couple of millimeters. There is no projecting ridge or fold above them such as occursin many Mniotiltidse. In Eucometisand Mitrospingus,the feathers come dose up to the posteriorend of the nostril. Tongue. The tongue ends in two points, one on each side; there is a series of fine teeth or serrations on each side, each tooth beinglarger than its proximalneighbor, so that the terminal tooth is the largest. Sometimesthe two terminal teeth on each sideare of equalsize. This sort of a tongueoccurs in many Passer- inc birdswhich have no closerelationship, as the Catbird and the Scarlet Tanager. It is not thereforein any way distinctive. Pterylosis. The head is fully leatheredwith no specialapteria nor is there any unusualarrangement of feathersin longitudinal series. The upper cervical tract is narrow at first but becomes broaderand moredensely feathered between the shouldersand then becomesnarrow again before joining the dorsaltract. Thisenlarge- ment of the cervicaltract is the only characteristicfeature of the generalpterylosis. It hasnot beenreported for any otherPasser- ine bird sofar as I know,but a somewhatsimilar arrangement is foundin tkekingfishers. The cervical tract anterior and posterior to this enlargementis, in Rhodinocichla,only three or four feathers wide but the enlargementis six featherswide. The rhombicdorsal saddle is well-marked and symmetrical and resemblesthat of Piranga erythromelasand many other Passeres. The pterylosis of the lower surfaceshows no peculiaritiessave that the ventral tracts are unusuallyshort, narrow and ill-defined. The primaries are rather short and the secondarieslong, giving the wing the short, roundedshape characteristicof the genus. The really remarkablefact hereis the shortnessof the eighthand ninth primaries. In the Mniotiltidse (with few exceptions)and in most tanagers,the eighth primary is one of the longestand the ninth is little shorterand is longerthan the fourth, but in Rhodinoclchla the eighth and ninth primariesare the shortestand the ninth is evenshorter than the secondaries.A very similararrangement of primariesis howeverfound in Mitrospinguswhich has the ninth primaryshortest and the eighth only a little longer,though it exceeds the first. In Eucometis,the ninth is longerthan the first and second,while the eighth is not much shorterthan the fifth, sixth Vol.1913 XXX] J CLARK,The Par•a•r•a Thrush-Warbler. 13

and seventh. In Piranga,the wingis pointedby the ninth, eighth and seventh primaries, the other extreme from Rhodi•ocichla. The latter hasnine secondaries and a quintocubitalwing of course. There are twelve rectriceswhich are successivelyshorter from the middle pair outward. Thesefeathers are notablefor their breadth and softness;it is interestingto find the tail feathersof Mitrospin- gus similar. Alimentary Canal. The arrangementof the intestine and the appearanceof the entire alimentary canal is so similar to that of severalother Passerine birds examined, Dumetella, Piranga, Seiurus, that no distinctive characters were found. The contents of the stomachwere examinedin two specimensand while much of the material wasunrecognizable to my untrainedeye, three itemswere determined; beetlesof at least four species,one of which was a curcullo; seeds,of which the most commonwas the hard gray acheneof somesedge; large, irregular grains of sand, with rounded angles and of a bright ochre color. This combinationseems to showclearly that the birdsare chieflyground feeders. Palatine Region. The bony palate of Rhodinocichlayields what seemsto me the bestindication of its relationships. If onecompares this part of the skull in tanagersand in woodwarblers, two points of differenceare shownwhich seemto be important. In the tanagers, the palatine processesare long and well developedwhile in the wood warblers they are short and rudimentary. In the tanagers, the maxillo-palatinesare parallel for a short distance, about the lengthof the inflatedportion, but in the warblersthey are parallel for a considerabledistance and the increasingdiver- genceis lessmarked. Parkeri examinedand figuredseveral species of each family, and I have examinedPiranga erythromelasand Seiu- rus noveboracensisin addition, and these differenceswhile not ex-

• Parker, W. K., 1878, Trans. Zool. Soc. London, vol. 10, pp. 251-314, pls. 46-54. Compare especially figs. 1-6, pl. 46 and figs. 1-3, pl. 48. It may be remarked in passing that Parker makes no reference to a "secondary palatine process" in Piranga, although he examined Pirar*ga rubra. Shufeldt figured these processes as found in Habia (Auk, Vol. 5, p. 439; 1888) and gave them a name. I find them very conspicuous in Piranga erttthromelas and on consulting Lucas' figure (Proc. U'. S. ]•lat. Mus., vol. 18, p. 505; 1895) it will be seen that he indicates them although he makes no reference to them. There is no trace of them in Seiurus or Dumetella, nor in Rhodlnocichla. They appear to be associated with a strongly eonlrostral beak, though Shufeldt says they are lacking in Coccothraustes. 14 CLA•K,Tile Pad, area Thrush-Warbler. [Jan..Auk

traordinary seem fairly constant. They show more clearly in specimensthan in Parker's figures, especiallythe position of the maxillo-palatines. Now in both these particulars, Rhodinocichla is a tanager. The palatine processesare well-developedthough not as slenderas in Piranga, nor asspike-like as Parker figuresthem for Tanagra. The maxillo-palatinesare almost exactly like those of Pira•ga erythromelasboth in form and position. Sternum. Someyears ago Shufeldt (1888, Auk, vol. 5, p. 442) made the statement that true tanagers"have an osseousbridge. extendingacross the top of the manubrium to the anterior margin of the body of the sternum." He adds that it is absentin Icteria and in such Fringillldrc as he had examined. This seemssuch a trivial characterand so unlikely to be constantor clearly shown, that I was inclinedto give little heed to it. Nevertheless I made somedissections with the result, surprisingto me, of finding this osseousbridge very clearly definedin Pira•ga and totally wanting in Dumctclla,Habia, Passerellaand Sciurus. I am forcedto believe therefore that it is a character of no little importance,and it is interestingto find that in Rhodi.•ocichla,this osseousbridge is as clearly shown as it is in Pira•g,. Conclusions.In the British Museum catalogue(1881), Sharpe callsRhodinocichla the "rose-breastedwren" and placesthe genus in the Miminrc near Harporhynch•s, remarking that it is one of severalgenera which appear to connectthe mocking-thrushesand the wrens. It was not until 1901 that Ridgway called attention to the important fact that there are only nine developedprimaries in Rhodinocichlaand that it must thereforebelong in someone of the nine-primariedgroups. He placedit finally at the end of the Mniotiltidrcwith the commentthat althoughit "is very aberrant as a memberof the Mniotiltid•e, I do not know where elseto place it." When Mr. Bangsplaced the alcoholicspecimens in my hands, he calledmy attention to the resemblanceto Mgtrospgng•sand sug- gestedthe possibilityof Rhod•'nocichlabeing a tanagcr. The evi- dence which I have presented seemsto me to justify the belief that this suggestionhas revealedthe probablerelationships of the genus. The structure of the bony palate and of the sternum are characteristicallytanagrinc, while the wing and tail sho•va close ßrelationship to Mitrospingus,which has quite generally been re- Vol.1913 XXX'] 3 EVERMANN,Birds New to thePribilof Islands. 15 gardedas a tanager. It is true that the bill is not typically tana- grinebut here againthere is a doseresemblance to Mitrospingus. It is a natural conclusion therefore that l•hodinoclchla is to be regardedas a tanager which has becomemore or lessspecialized for a particular manner of life. As the stomachcontents indicate a groundfeeder, it may be that that method of finding its living has beenthe factor associatedwith its specialization.

EIGHTEEN SPECIES OF BIRDS NEW TO THE PRIBILOF ISLANDS, INCLUDING FOUR NEW TO NORTH AMERICA2

BY BARTON WARREN EVERMANN.

W•Tn the appointment of a naturalist in the fur-seal service July 1, 1910, and the organization,in the Bureau of Fisheries,of the AlaskaFisheries Service, July 1, 1911,the Bureauat oncebegan the formation of plans for a comprehensiveand thoroughstudy not only of the life history of the fur seal but also of the scientific managementand conservationof the fur-seal herd that has its breedinggrounds on the Pribilof Islandsin BeringSea. The plan is broadin its scopeand contemplatesa thoroughstudy of all the speciesof animalsand plantsfound on or about thoseislands. Dr. Walter L. Hahn, at that time head of the departmentof biology in the state normal schoolat Springfield,South Dakota, was ap- pointednaturalist in the summerof 1910. He arrivedat St. PaUI Island August24 and immediatelyentered upon his dutieswith an energy and intelligencewhich could scarcelybe excelled. His untimely death on May 31, 1911, froin exposurein the ice-cold water of the villagelagoon, resulting from the capsizingof a boat, was a severeloss to the fur-seal serviceand to biologicalscience. During his few months on St. Paul Island Dr. Hahn, from the

Published witch t•he permission of t•he iT. S. Commissioner of Fisheries.